GCL-R-Scripts

Update 2023-09-11: This repository is no longer actively maintained. These R functions have been superceeded by the GCLr package.

A repository for all GCL R Scripts. This repository is meant to store and provide version control for all shared GCL scripts. New functions should be developed and included as necessary. Older functions that previously existed in the "temp" folder will be migrated into this repository over time, so all can use them. GitHub is a collaborative tool and is meant to store not only our code, but to also allow us to keep all R script related issues and discussions in one accessible place.
Files were initially moved off of the V: drive on 2/29/16.

Initial cloning of repository

• Download RStudio
• Dowload Git
• Create a GitHub account
• Enable version control in RStudio via Tools --> Global Options --> Git/SVN
• Open your git-bash.exe ("C:\Program Files\Git\git.bash.exe") and follow code below, changing to your username and e-mail. Note that the "Repository URL" does not change to your username as it lives in krshedd's account.
  • Configure Git with your GitHub account
    • Set your username to your GitHub username git config --global user.name "krshedd"
    • Set your e-mail address to what you use for GitHub git config --global user.email "[email protected]"
    • Verify changes git config --global --list
  • Set up SSH key for RStudio to communicate with GitHub
    • Change directory to where you want to save key cd /c/Users/krshedd
    • Make directory to save key in mkdir ".ssh"
    • Create key ssh-keygen -t rsa -b 4096 -C "[email protected]"
    • Enter file in which to save the key /c/Users/krshedd/.ssh/id_rsa
    • Create passphrase for key
    • Re-enter passphrase for key
    • Open ssh-agent eval "$(ssh-agent -s)"
    • Add key to ssh-agent ssh-add /c/Users/krshedd/.ssh/id_rsa
    • Enter passphrase for key
    • Copy SSH key clip < /c/Users/krshedd/.ssh/id_rsa.pub
    • Follow instructions to add SSH key to your GitHub account online
    • Test your SSH connection ssh -T [email protected]
    • Are you sure you want to continue connecting yes
    • After confirmation, close git.bash.exe
• Initial clone of GCL-R-Scripts repository on GitHub to your local machine
  • Open a new project in RStudio via File --> New Project... --> Version Control --> Git
    • Repository URL: https://github.com/krshedd/GCL-R-Scripts.git
    • Project directory name autofills
    • Create project as subdirectory of: wherever you want on your C: drive (i.e. "C:\Users\krshedd\Documents\R")
    • Close project; you now have a local copy of all files from the GCL-R-Scripts repository.
• As a final step, you need a Functions.GCL.R R-script that will source all of the R-scripts from your local directory. Note that you will need to change directory to the path to your local "cloned" repository.

#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# This function sources GCL functions from my C drive (GitHub clone)
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
directory <- "C:/Users/krshedd/Documents/R/GCL-R-Scripts"
functions <- list.files(path = directory, pattern = "\\.R$", full.names = TRUE, recursive = FALSE, ignore.case = TRUE)  # read in all .R scripts
EMPTY <- sapply(functions, source)
rm(directory, functions, EMPTY)
print(as.vector(lsf.str()))

• Once you have created the above Functions.GCL.R, you will source to it directly to access all GCL-R-Scripts
source("C:/Users/krshedd/Documents/R/Functions.GCL.R")

Normal usage

Best practice is to pull frequently (i.e. beginning of each day).

• Open your GCL-R-Scripts.Rproj in RStudio
• In the upper right panel select Git
• Confirm that you are on the master branch (upper right)
• Select the blue down arrow for Pull. A window will indicate whether you are already-up-to-date or inform you of changes since your last pull
• Close GCL-R-Scripts.Rproj
• You now have the most up to date version of all R scripts on your local C: drive

Releases

Periodically, after large changes to the GCL-R-Scripts repository, a specific commit will be tagged as a release. Each release will have a version number and a thorough set of release notes indicating:

1. New Functions
2. Function Updates
3. Bug fixes
   Release notes serve to update you periodically of major changes so you don't have to go through each commit specifically to note changes.

Branches

The GCL-R-Scripts repository has two main branches: master and develop. The thought is to loosely follow this successful Git branching model. The master branch is what we should all be pulling from and sourcing to. Ideally, the develop branch is where new functions are added and changes are made to existing scripts between releases. When it is time for a new release the develop branch is merged back to the master branch once code has been vetted and tested. If there are bugs in a release that need immediate attention, a temporary hotfix branch should be created from the master to address the bug.

Issues

One of the most powerful features of GitHub is tracking Issues. When someone has a problem with one of our functions, finds a bug, or wants a new feature, it is best to create a Issue. This allows us to:

1. Document the problem a user had with a time-stamp,
2. Discuss potential solutions,
3. Come to a consensus solution,
4. Assign a person to resolve the issue, and
5. Write code that solves the problem and closes the Issue
   This is a much better solution than e-mailing Jim or individual users. This allows everyone to know what the problem is, who had it, and if it has been fixed. Issues can be closed by commits, and should not be closed until code is written to solve the Issue.

Sourcing scripts from Remote Desktop

If you need to source R scripts, but do not have access to your local C: drive (i.e. when you are on the remote desktop servers), you can source scripts individually by their unique https url. All R scripts can be accessed by their raw https url which follows the formula "https://raw.githubusercontent.com/*username*/*repository*/*branch*/*filename*", for example https://raw.githubusercontent.com/krshedd/GCL-R-Scripts/master/AttributesToIDs.GCL.r.
Another way to find the raw https url is to navigate within the repository on GitHub to the function, select Raw, and copy the url.
Once you have the url for the function you want to source there are two options to source it. The first one is easiest (less code), but the second one will always work (requires the package "RCurl").

source(https://raw.githubusercontent.com/krshedd/GCL-R-Scripts/master/AttributesToIDs.GCL.r)`

while(!require(RCurl)){install.packages("RCurl")}
eval(expr = parse(text = getURL("https://raw.githubusercontent.com/krshedd/GCL-R-Scripts/master/AttributesToIDs.GCL.r", ssl.verifypeer=FALSE) ))

QC

As of release QC For Everyone v1.4.0 the standard QC R script run by project leaders will work for both SNPs and uSATs and is available on the master branch. This is a separate script all of its own, not a full-blown function. Weird things always seem to happen during QC, so rather than have a black box function, it was decided to create a dedicated script that project leaders would then run line by line so that they can pinpoint problems if/when they arise. The idea is that when you need to perform a QC, you will do a fresh pull from the master branch in order to have the latest version of QC.R in your remote repository (i.e. C:\Users\krshedd\Documents\R\GCL-R-Scripts). You will then copy/paste this file to the QC project folder (i.e. V:\Lab\Genotyping\SNP Projects\Sockeye\Project S161 Chignik Inseason 2016\QC), modify the Arguments, and run the script from there. You should not have to modify anything past GO!, but if the function bombs on you, you will at least know where you are. This arrangement will also allow you to save your project specific QC script on the V drive when you are done.